Student Victimization: The Impact of Educational Systems
on School Violence in 37 Nations
by Tracy Mason

School violence is an issue of international concern, yet there are no comprehensive studies of the cross-national factors that cause school violence–until now that is.

David Baker, professor of education and sociology, teamed with Gerald Le Tendre, associate professor of education, and Motoko Akiba, a Penn State graduate student in education theory and policy, to develop an understanding of how macro-sociological factors–such as the presence of shadow education or overall levels of academic competition–are important links in understanding the correlations and causes of violence in schools.

"In this paper, we utilized a little-analyzed section of the Third International Math and Science Study (TIMSS) data," Baker said. "We wanted to 1) explore how much school violence there is in the world; 2) determine whether or not previous theories of juvenile delinquency and school violence held up in cross-national analysis; and 3) test whether factors related to the educational system itself were associated with levels of school violence," Baker added. 

The 1994 TIMSS data used for this research contained multiple measures of violence: two types of violent behaviors based on student reports, two types of delinquent behaviors based on teacher reports and six types of violence behaviors based on school reports.

"We found that violence is widespread in schools around the world," Baker noted.

Of the U.S. students surveyed, 25.7% stated that they got hurt or were threatened to get hurt at least once a month. At the high end of the scale, about 75% of the students in Hungary and 67% of Romanian students felt the same while more than 6% of students surveyed from Denmark felt the same. In that category of questioning, the worldwide sample mean was 27.8%

When teachers were asked "How much is your teaching limited by threats to your personal safety or student’s safety?" only 7.6% of the U.S. respondents responded positively that a risk of violence existed. This is significantly lower than the mean response of 13.7%.

"The U.S. is not very high on the violence scale in each category of questioning," shared Baker. "We’re not trying to downplay the seriousness of events like Columbine; yet overall, American schools are not significantly more violent than schools in other countries throughout the world," Baker said. "We do tend to ‘be more violent’ than other industrialized nations though," he added.  

"Previous theories do not adequately explain the structural causes of violence; a combination of poor-quality public schooling combined with strong academic competition is significantly associated with higher levels of student victimization," Baker said. "When you put kids in special remedial classes with little resources, you’re just asking for trouble," Baker warned.

One surprise finding notes that schools that are more effective at teaching math tend to report less acts of violence. Schools that have a spread or variation of math scores strongly produce more violence (with some classroom to classroom variation).

In addition, analysis results show that national patterns of school violence are not related to general patterns of violence or social disintegration, and that school-related factors appear to be more powerful predictors of system-wide levels of violence.

Contact: Jeff Deitrich -- 814-863-2216 FAX: 865-0555
http://www.ed.psu.edu